Sea Ice

What are scenarios? In our context a scenario is one of many possible pictures of
the future. A scenario is not meant to forecast the future but to set the reader
thinking about possible futures. Usually a scenario process provides three to five
scenarios from the very pessimistic to the very optimistic range. These scenarios
are then used to develop a strategy that is flexible and robust enough to work for
most of the scenarios and contains an ‘exit strategy’ for the worst case.
We distinguish between narrative scenarios and explorative scenarios. Narrative
scenarios are creative stories about the future that are not necessarily based
on research. Whereas, explorative scenarios are based on extensive research and
analysis of the data. ScenLab that supports the building of explorative scenarios.
A good scenario not only represents a certain picture of the future but explains
how the present will evolve into this future. It is required that the developments
towards this future are consistent and plausible. In narrative scenarios this task
can be easily accomplished by doing ‘backwards engineering’, i.e. the scenario is
written first and then how it happened that the present developed in this certain
future.
Developing explorative scenarios is a more difficult task. Ideally, we the goal
is to find, with as little prejudice and assumptions as possible, of all theoretically
possible developments the ones that are most likely.
A standard scenario describes the ‘smooth’ (continuous) development from the
present to a possible future. In order to account for discontinuities it is possible
to perform a disruptive events analysis by using so-called wild cards. Wild cards
are highly unlikely events with a major impact on future developments.
The process of building explorative scenarios consists of five stages. In the first
stage the field of research is determined and the time and scope of the analysis are
set.
In the second stage the key factors and their respective future projections are
found. A key factor is a driving force of the researched field. Its development
will strongly influence the development of the whole field. Key factors can be
anything from environmental influences, over economic and social developments,
to the behavior of stake-holders and competitors, and wild cards. Each of these key
factors can usually develop in more than one way. Therefore, future projections
are assigned to each key factor. Future projections describe possible developments
of a key factor. Depending on the required accuracy of the scenarios and the
complexity of the field under investigation usually between 10 and 35 key factors
with 1 to 5 respective future projections are identified.
The crucial question for stage three is, which future projections of the key
factors are likely to occur in the same consistent and plausible future? Off course
it would be easy to simply pick a set of future projections (a so-called projection
bundle) and check if they make a consistent and plausible future. However, this is
not very accurate and very likely to miss crucial combinations of future projection.
To avoid this problem the consistency and robustness analysis are carried out, these
are described in detail in the following section.
The consistency and robustness analysis provide the data for stage four – the
identification of raw scenarios. This is done by analyzing the data with the goal to
find the best candidates for scenarios out of the, usually very large, list of projection
bundles. The data analysis tools provided by ScenLab are multi-dimensional
scaling, clustering and distribution plots. For more information see chapter ??.
The identified future projections build the skeleton for the final scenarios written
in stage five. Sofar, we only have a accumulation of data that characterizes
likely future developments. The task is to write good, easily understandable pictures
of the future using this data.
Keep in mind that almost the whole process of building explorative scenarios is
based on human intelligence and expert knowledge. A software tool is not able to
actually write scenarios, it is only able to support the expert user in finding ‘good’
scenarios. There is a lot of ambiguity in the process due to the fact that different
evaluations of the data lead to different results. Therefore, it is highly advisable
to perform the scenario process as a team discussing the selection of key factors,
future projections, the evaluation of the input and output data as extensively as
possible.